ECG.jpg

Cardiovascular risk detection

Continuous daily detection and observation for heart arrhythmias that are difficult to detect with a single medical test. No complex equipment is required, which means taking day-to-day measurements is more convenient and simple. Can be used in collaboration with healthcare units to detect early signs of heart disease risk through cloud intelligence monitoring for early medical treatment and further examination arrangements.

Acer Healthcare Inc. and partners have developed an algorithm for arrhythmia. The algorithm can be used in conjunction with an Acer Leap Ware wearable device, so that patients may monitor their own day-to-day health. Physiological signals are continuously recorded using the Acer wearable device. Patients can use the health management platform to view their heart rate, graphs for heart rate distribution and waveform, and other AI arrhythmia analysis results. Results on the health management platform can be used to inform patients if any arrhythmia or cardiovascular-related risk signals are present.

心血管-2.jpg

Abdominal ultrasound —
organ and lesion recognition technology

Instant scanning for timely calculations to immediately detect lesions

Recognition accuracy

95%

Liver,
gall bladder, kidneys

90%

Gallstones, cholecystitis

80%

Liver cysts, gallbladder polyps

Acer Healthcare Inc. and partners have developed ultrasound-imaging-assisted diagnosis technology. Combining AI technology with ultrasound diagnostic equipment, physicians are able to make timely interpretations of abdominal ultrasound images and quickly mark abdominal organs and possible lesions on the detection screen following an abdominal scan. Photographs of suspected lesions can also be captured, providing physicians with the ability to make further diagnoses and the means to better understand the condition of various abdominal organs.

Echocardiography —

automatic left ventricular ejection fraction measurement

Chamber Recognition Accuracy

98%

A2C View
A4C View

​Other View

Segmentation Accuracy

91%

A2C Endocardium
A4C Endocardium

Mean Absolute Error
Of Ejection Fraction

±1.2

Biplane Simpson’s

Acer Healthcare Inc. and partners have developed the technology of automatic left ventricular ejection fraction measurement. This technology is using artificial intelligence technology to improve the efficiency and accuracy by identifying the chamber views, image segmentation and modified Simpson’s method. And the technology has been successfully developed on mobile devices in real-time measurement of cardiac blood volume and the ventricular ejection fraction which can also used on wireless handheld ultrasound scanning devices.

Harmony Referral System

Convenient and fast access to patient dataset, with AI-assisted screening and clinical analysis

Harmony Referral System improves the management of your patient exam data by allowing faster access, seamless EMR (Electronic Medical Records) integrations and connection to every instrument. HRS provides AI-assisted early screening and clinical data analysis to allow physicians to make diagnoses more accurately, and can also conduct multi-party consultations and provide telemedicine services through the cloud service.

Link: https://www.topconhealth.com/

Early Renal Function Impairment Detection

Related Paper: Journal of Medical Internet Research (submitted in 2020). A Deep Learning Model for Detecting Early Renal Function Impairment Using Retinal Fundus Images: Model Development and Validation Study (IF 4.945) 

Paper Link: https://preprints.jmir.org/preprint/23472/accepted
 

Influenza Forecasting Model

Related Paper : Journal of Medical Internet Research (2020). Applying Machine Learning Models with An Ensemble Approach for Accurate Real-Time Influenza Forecasting in Taiwan: Development and Validation Study (IF 4.945)

Paper Link : https://www.jmir.org/2020/8/e15394/

Multimorbidity Frailty Index

Related Paper : Journal of Medical Internet Research (2020). Comparisons Between Hypothesis- and Data-Driven Approaches for Multimorbidity Frailty Index: A Machine Learning Approach (IF 4.945)

 

Paper Link : https://www.jmir.org/2020/6/e16213/